基于用户的协同过滤个性化音乐推荐系统
摘 要
互联网发展到如今已经完完全全的改变了的生活方式，融入了日常生活，包括交流，出行，消费，娱乐等。与此同时，音乐数据也在与日俱增的变化着。用户在访问一个音乐网站时，如何能快速的找到自己想要享受的歌曲呢？个性化音乐推荐系统可以做到。
课题做了个性化推荐系统，后端使用个性化推荐算法构造，前端使用spring+ssm框架搭建了个性化音乐推荐系统。系统数据库使用了关系型数据库MySQL和大数据数据库。前端收集过用户行为数据后传到后端使用基于用户的协同过滤算法来推荐出用户可能喜欢的音乐。设计主要完成了从网易云音乐门户网站上爬取数据（音乐信息，歌手信息等），并在获取数据后对数据进行清洗过滤等操作后保证了数据的有效性，将爬取到的六千多条数据保存到数据库后，采用基于用户的协同过滤算法推荐用户可能喜欢的音乐。
人们的需求随着数据量的不断增大在不断的增加，而且人们对音乐的追求也随着也随着这些不断的增多。个性化推荐系统的产生顺从了时代的发展，在不远的将来，个性化推荐系统必能大放异彩。
关键词：音乐推荐；基于用户的协同过滤；数据爬取；数据处理；推荐系统。
Personalized music recommendation system
Abstract
The development of the Internet has now completely changed the way of life, integrated into daily life, including communication, travel, consumption, entertainment and so on. At the same time, music data is changing day by day. When users visit a music website, how can they quickly find the song they want to enjoy? Personalized music recommendation system can do it.
The subject made a personalized recommendation system, the back-end was constructed using a personalized recommendation algorithm, and the front-end used the spring+ssm framework to build a personalized music recommendation system. The system database uses the relational database MySQL and the big data database. The front-end collects user behavior data and transmits it to the back-end to use the user-based collaborative filtering algorithm to recommend music that the user may like. The design mainly completed the crawling of data (music information, singer information, etc.) from the Netease cloud music portal website, and after the data was obtained, the data was cleaned and filtered to ensure the validity of the data. After multiple pieces of data are saved to the database, a user-based collaborative filtering algorithm is used to recommend music that the user may like.
People's needs are increasing with the increasing amount of data, and people's pursuit of music is also increasing with these. The generation of personalized recommendation system is in line with the development of the times. In the near future, the personalized recommendation system will definitely shine.
Keywords: music recommendation; user-based collaborative filtering; data crawling; data processing; recommendation system.
